agent-browser

To automate browser interactions for web testing, use a CLI to open URLs, generate page snapshots, and execute actions like click, fill, or screenshot. This enables reliable form filling and data extraction across multiple websites.

Ref-based element targeting uses references like @e1 or @e2 generated from page snapshots to interact with specific page elements. This ensures stable browser automation by clicking or filling elements using consistent identifiers.

You extract data and capture screenshots during web automation by executing the get and screenshot operations via the CLI. These actions gather information and record visual state directly from the browser session.

Yes, you can manage optional session state across multiple websites. This maintains browser context and state when performing automated demonstrations, QA workflows, and data gathering tasks across different sites.

Available browser actions for QA workflows include open, snapshot, click, fill, get, screenshot, and wait. These CLI operations support comprehensive web testing, data extraction, and automated demonstration tasks.

No, you do not need any external dependencies to run browser automation for testing. The skill operates independently without requiring additional components or external libraries to execute its CLI commands.